Loss of Fragile X mental retardation protein (FMRP) due to mutations in the Fmr1 gene causes Fragile X syndrome (FXS), the most common form of inherited mental disability and the leading genetic cause of autism. Despite two decades of intensive studies characterizing FMRP functions at synapses, the molecular basis of FXS remains poorly understood. FMRP is thought to function primarily as a regulator of protein synthesis in dendrites, and research to date on FXS has concentrated on the postsynaptic effects of FMRP loss leading to altered long-term synaptic plasticity (LTP). While LTP is thought to play important roles in learning and memory, short-term plasticity (STP) is widely believed to control other essential neural functions such as information processing, working memory and decision making. STP dysregulation may thus play a significant role in the cognitive impairments in FXS. However, STP dysregulation in FXS has received little attention and is poorly understood. Moreover, whether FMRP plays a role in synaptic mechanisms controlling STP remains largely unknown. Our recent studies revealed that loss of FMRP causes marked STP defects and abnormal information processing in excitatory hippocampal synapses. We further demonstrated that FMRP loss causes abnormal increase of a major calcium-dependent form of rapid presynaptic enhancement, known as augmentation, and that the calcium influx in presynaptic neurons is also increased. We therefore hypothesize that altered presynaptic calcium dynamics represents a major underlying cause of STP defects in the absence of FMRP. Importantly, our results indicate that at least some of the underlying mechanisms of these defects have a cell-autonomous presynaptic origin and arise from a novel FMRP function that is not related to its traditional role in protein translation. We propose to combine electrophysiological and imaging approaches with pharmacology and molecular biological tools to (i) determine how loss of FMRP alters calcium dynamics and STP; (ii) Examine the functions of FMRP mediating these defects; and (iii) Determine the impact of synaptic abnormalities associated with FMRP loss on computations performed by canonical neural circuits. We anticipate that these studies will provide fundamental new insights into the function of FMRP in synapses and a novel way to approach synaptic dysfunction in FXS.